Questions?
Call 603 659 5335 or email me

The long-awaited HUMAN 81 DK is now available.
Best viewed in "landscape" mode.

Ferrofluid Lifetime in Use

Draft

Return to Essay Topics

Ferrofluid is a special magnetic liquid that we use in some speaker parts - almost all tweeters, some midranges, and a few woofers - to achieve two goals. One is improved cooling, where it provides a better thermal coupling of the voice coil with the magnet structure that can then act as a heat sink, and the other is mechanical damping of the moving assembly. The cooling is generally the primary goal, as it can increase power handling markedly, and also reduce dynamic compression due to heating in the voice coil. A third function that the Ferrofluid does serve to some degree is in centering the voice coil in the gap. It can do this quite well, but should not be the only way that is ensured.

Ferrofluid does not "dry out" over time, and neither should it "leak".

It is a permanent emulsion of (formerly natural, now synthetic) oil and magnetic particles, with carefully chosen and specified viscosities and magnetic densities.

For the oil to "dry out" some of it would have to have a pathway to escape, and it would have to be unstable at normal operating temperatures. This material was invented and developed to lubricate bearings in space - under zero gravity and in a hard vacuum. For it to "leak", likewise, it would have to first escape the magnetic field it is trapped in, and then manage to work its way out of the device in question.

None of these things happens in the normal use and playing of a home (or car, etc.) loudspeaker system.

What does happen, however, is that speakers are overdriven, whether by accident or abuse, and the Ferrofluid can be heated beyond its operating design range. This can result in some of it vaporizing, but what is much more likely is it getting burnt - much like automobile engine oil that is left in a red-hot turbocharger.

Also, mechanical splattering can occur. This and/or the boiling off can result in some of the liquid escaping the gap and potentially being wicked through a porous tweeter dome (for example). At this point it is no longer where it was designed to be. Again, this requires operation outside of the design limits to happen (if the speaker was properly designed).

Both of these processes - and it is far more the burning, the coking, of the oil than the Ferrofluid escaping - result in less of the magnetic oil being in the gap where it belongs.

Note that both of these decreases in functionality are the gradual result of overdriving the speaker system - either due to repeated accidents, abuse, or poor initial design. Each time this happens without actually destroying the driver, a little less Ferrofluid is left behind to do its job.

Eventually, the now-reduced quantity of Ferrofluid ceases to offer enough protective (or damping) properties to cope with these events.

In its absence, the drivers in question that would have died of these operating conditions far earlier may finally succumb to this "normal operating level" and fail due to burned-out voice coils.

Ferrofluid does not degrade or "dry out" or anything like that in normal use. I have been taking tweeters apart for various reasons that were made as long ago as 1975, and unless they were overdriven (the now-dead tweeters), it is always perfectly fine (much like every crossover capacitor I have ever tested that was not obviously physically damaged). It certainly does not cause tweeters to die - it is the opposite. It helps them survive.

(That does not prevent people from claiming it does and offering to "fix" this problem.)

The reason people find it cooked is that they are opening up parts with issues - damaged parts - and part of the problem is that the Ferrofluid has been burnt up.

The tweeters in question being damaged by their owners has nothing to do with Ferrofluid at all.

Along with the occasional accident or abuse, people will tend to keep playing a system louder and louder as long as there is no obvious distortion. Ferrofluid lowers distortion at high levels, so this can result in a sense of "normal operating levels" actually being outside what the designer ever envisioned.

The post-mortem of most blown/dead/burnt out tweeters is going to show some issues with the Ferrofluid. But it was not the Ferrofluid that caused the problem, the owner did, and the Ferrodfluid was collateral damage, burnt or charred, "coked" as they say, in the process of destroying the tweeter voice coil (which may have taken many insults before the injury finally was too much).

People are not good at processing this sort of information - that a blown tweeter exhibits "dried out", coked and burnt Ferrofluid - and tend to leap to the wrong conclusion and repeat it as a deep revealed truth. Especially if someone can make some money from it.

Examining an old driver's internal condition, if possible, should not result in any harm. As long as it is kept clean, and exactly accurate part alignment is ensured somehow, such an examination might indeed show signs of abuse and overheating that can possibly be remedied by incredibly careful cleaning and replenishment with exactly the right amount of the right viscosity and

Note that, specific to the products I make and brands I specialize in, the first generation EPI/Epicure tweeters - most of the ones with Masonite faceplates and the three asymmetrical mounting holes - do not have Ferrofluid in them. They do, however, have four tiny pieces of decaying foam that I think acted to help center the voice coil during assembly, and some silicon grease intended to help with heat transfer (cooling).

PS, I thought that "Ferrofluid" was a trademark of Ferrotec (USA) Corporation (formerly Ferrofluidics), but they do not claim this on their web site.

* * *
Top - Contact -
General Topics (Home)
New Speakers - Owner Support - DIY Projects
EPI and Epicure - Genesis Physics -
All Other Speaker Repairs

© Copyright - All Rights Reserved

Al Major Credit and Debit Cards Accepted

7 Kelsey Road, Lee, New Hampshire 03861

Prices can change and specifications will improve without notice

Load a printer-friendly version of this page